| 1 |
#! /usr/bin/env python3 |
| 2 |
# This script generates Lib/re/_casefix.py. |
| 3 |
|
| 4 |
import collections |
| 5 |
import sys |
| 6 |
import unicodedata |
| 7 |
|
| 8 |
SCRIPT_NAME = 'Tools/build/generate_re_casefix.py' |
| 9 |
|
| 10 |
def update_file(file, content): |
| 11 |
try: |
| 12 |
with open(file, encoding='utf-8') as fobj: |
| 13 |
if fobj.read() == content: |
| 14 |
return False |
| 15 |
except (OSError, ValueError): |
| 16 |
pass |
| 17 |
with open(file, 'w', encoding='utf-8') as fobj: |
| 18 |
fobj.write(content) |
| 19 |
return True |
| 20 |
|
| 21 |
re_casefix_template = f"""\ |
| 22 |
# Auto-generated by {SCRIPT_NAME}. |
| 23 |
|
| 24 |
# Maps the code of lowercased character to codes of different lowercased |
| 25 |
# characters which have the same uppercase. |
| 26 |
_EXTRA_CASES = {{ |
| 27 |
%s |
| 28 |
}} |
| 29 |
""" |
| 30 |
|
| 31 |
def uname(i): |
| 32 |
return unicodedata.name(chr(i), r'U+%04X' % i) |
| 33 |
|
| 34 |
class hexint(int): |
| 35 |
def __repr__(self): |
| 36 |
return '%#06x' % self |
| 37 |
|
| 38 |
def alpha(i): |
| 39 |
c = chr(i) |
| 40 |
return c if c.isalpha() else ascii(c)[1:-1] |
| 41 |
|
| 42 |
|
| 43 |
def main(outfile='Lib/re/_casefix.py'): |
| 44 |
# Find sets of characters which have the same uppercase. |
| 45 |
equivalent_chars = collections.defaultdict(str) |
| 46 |
for c in map(chr, range(sys.maxunicode + 1)): |
| 47 |
equivalent_chars[c.upper()] += c |
| 48 |
equivalent_chars = [t for t in equivalent_chars.values() if len(t) > 1] |
| 49 |
|
| 50 |
# List of codes of lowercased characters which have the same uppercase. |
| 51 |
equivalent_lower_codes = [sorted(t) |
| 52 |
for s in equivalent_chars |
| 53 |
for t in [{ord(c.lower()) for c in s}] |
| 54 |
if len(t) > 1] |
| 55 |
|
| 56 |
bad_codes = [] |
| 57 |
for t in equivalent_lower_codes: |
| 58 |
for i in t: |
| 59 |
if i > 0xffff: |
| 60 |
bad_codes.extend(t) |
| 61 |
try: |
| 62 |
bad_codes.append(ord(chr(i).upper())) |
| 63 |
except (ValueError, TypeError): |
| 64 |
pass |
| 65 |
break |
| 66 |
if bad_codes: |
| 67 |
print('Case-insensitive matching may not work correctly for character:', |
| 68 |
file=sys.stderr) |
| 69 |
for i in sorted(bad_codes): |
| 70 |
print(" '%s' (U+%04x, %s)" % (alpha(i), i, uname(i)), |
| 71 |
file=sys.stderr) |
| 72 |
sys.exit(1) |
| 73 |
|
| 74 |
mapping = {i: tuple(j for j in t if i != j) |
| 75 |
for t in equivalent_lower_codes |
| 76 |
for i in t} |
| 77 |
|
| 78 |
items = [] |
| 79 |
for i, t in sorted(mapping.items()): |
| 80 |
items.append(' # %s: %s' % ( |
| 81 |
uname(i), |
| 82 |
', '.join(map(uname, t)), |
| 83 |
)) |
| 84 |
items.append(" %r: %r, # '%s': '%s'" % ( |
| 85 |
hexint(i), |
| 86 |
tuple(map(hexint, t)), |
| 87 |
alpha(i), |
| 88 |
''.join(map(alpha, t)), |
| 89 |
)) |
| 90 |
|
| 91 |
update_file(outfile, re_casefix_template % '\n'.join(items)) |
| 92 |
|
| 93 |
|
| 94 |
if __name__ == '__main__': |
| 95 |
import sys |
| 96 |
main(*sys.argv[1:]) |